- Former Anthropic/OpenAI researcher Jacob Coxon resigned and warned that engineers are too heads-down to weigh AI safety, even saying some builders fear the tech "could kill us all" by the end of the decade.
- Anthropic CEO Dario Amodei urged slowing the pace of AI capability advances to allow safety measures to catch up — a step Coxon praised and said should include cooperation with Chinese firms.
- Lawmakers sounded alarms and urged safeguards, but political friction (including a canceled two-week House session) is complicating a swift regulatory response.
The news, distilled into what matters.
Get the full experience in the app — topics, comments, and audio summaries.